Võrdle meetodeid
Vaata valitud meetodeid kõrvuti; erinevad read on esile tõstetud.
| Selgitatav tugevdamisõpe× | Tähelepanu mehhanism× | |
|---|---|---|
| Valdkond | Süvaõpe | Süvaõpe |
| Perekond | Machine learning | Machine learning |
| Tekkeaasta≠ | 2018–2020 | 2015 |
| Looja≠ | Puiutta, E. & Veith, E. M. S. P. (survey); broader XAI community | Bahdanau, D.; Luong, M.T. |
| Tüüp≠ | Hybrid approach (RL + explainability methods) | Neural attention layer (encoder-decoder) |
| Algallikas≠ | Puiutta, E., & Veith, E. M. S. P. (2020). Explainable Reinforcement Learning: A Survey. In Machine Learning and Knowledge Extraction (CD-MAKE 2020), Lecture Notes in Computer Science, vol. 12279, pp. 77–95. Springer. DOI ↗ | Bahdanau, D., Cho, K. & Bengio, Y. (2015). Neural Machine Translation by Jointly Learning to Align and Translate. ICLR. link ↗ |
| Rööpnimetused≠ | XRL, interpretable reinforcement learning, transparent RL, explainable RL | Dikkat Mekanizması (Bahdanau / Luong Attention), dikkat mekanizmasi, neural attention, additive attention |
| Seotud≠ | 3 | 5 |
| Kokkuvõte≠ | Explainable Reinforcement Learning (XRL) augments standard reinforcement learning agents with methods that make their policies, decisions, and learned behaviors interpretable to humans. Rather than treating the policy as a black box, XRL produces post-hoc explanations or builds inherently transparent policies, enabling trust verification, debugging, and accountability in high-stakes automated decision-making. | The attention mechanism, introduced by Bahdanau, Cho and Bengio in 2015 and refined by Luong, Pham and Manning the same year, lets a sequence decoder dynamically learn which of the encoder's outputs to focus on at each step. Before the Transformer, it substantially improved machine-translation quality by freeing models from compressing an entire input into a single fixed vector. |
| ScholarGateAndmestik ↗ |
|
|